Acoustic Model Fusion for End-to-end Speech Recognition

TL;DR

This paper introduces a novel method of fusing an external acoustic model into end-to-end speech recognition systems to better handle domain mismatch and improve accuracy, especially for named entity recognition.

Contribution

The paper proposes integrating an external acoustic model into end-to-end speech recognition systems, addressing domain mismatch issues and improving recognition accuracy.

Findings

Achieved up to 14.3% reduction in word error rate.

Enhanced named entity recognition performance.

Demonstrated effectiveness across diverse test sets.

Abstract

Recent advances in deep learning and automatic speech recognition (ASR) have enabled the end-to-end (E2E) ASR system and boosted the accuracy to a new level. The E2E systems implicitly model all conventional ASR components, such as the acoustic model (AM) and the language model (LM), in a single network trained on audio-text pairs. Despite this simpler system architecture, fusing a separate LM, trained exclusively on text corpora, into the E2E system has proven to be beneficial. However, the application of LM fusion presents certain drawbacks, such as its inability to address the domain mismatch issue inherent to the internal AM. Drawing inspiration from the concept of LM fusion, we propose the integration of an external AM into the E2E system to better address the domain mismatch. By implementing this novel approach, we have achieved a significant reduction in the word error rate, with an impressive drop of up to 14.3% across varied test sets. We also discovered that this AM fusion approach is particularly beneficial in enhancing named entity recognition.